In the last three decade, there has been a marked increase in the prevalence of childhood obesity. Currently, 23% of preschoolers (2-5 years) in the United States are overweight or obese, a critical period to target in hopes of reversing this trend. Emerging consensus indicates that excessive, intrusive, overt control in feeding or a chaotic non-structured feeding environment are risk factors for childhood obesity. Targeting these behaviors has significant potential for obesity treatment. In this 3-year R03 application, we propose to investigate the efficacy of a Feeding Dynamic Intervention (FDI) for improving caregiver feeding practices, child eating behaviors, and child self-regulation of energy intake in the short term. In a pilot randomized controlled trial of 84 mothers with 3- to 5-year-old obese children, we will examine the effect of our FDI in comparison to a wait-list control group (WLG) on (i) maternal self-reported feeding practices; (ii) improving energy compensation (COMPX), decreasing Eating in the Absence of Hunger (EAH) and increasing mother-reported satiety responsiveness in their child. Finally, we will investigate an exploratory hypothesis that children in the FDI group, compared to children in the wait-list control group, will demonstrate a lower increase in body mass index (BMI) z-score over 6 months. At the completion of this study, we expect to know: (1) the effect of the FDI on maternal self- reported feeding practices for obese children; (2) the degree to which children can be taught to self-regulate their food intake via parent-directed intervention; and (3) the estimated effect size to enable us to calculate sample size requirements for a subsequent large-scale multicenter trial in an R01 application. Our long-term research goal is to develop effective and pragmatic interventions that enable children to improve their self-regulation of energy intake and maintain a healthy weight.